includes/content/ContentHandler.php

   1 <?php
   2 /**
   3  * Base class for content handling.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @since 1.21
  21  *
  22  * @file
  23  * @ingroup Content
  24  *
  25  * @author Daniel Kinzler
  26  */
  27
  28 /**
  29  * Exception representing a failure to serialize or unserialize a content object.
  30  *
  31  * @ingroup Content
  32  */
  33 class MWContentSerializationException extends MWException {
  34 }
  35
  36 /**
  37  * A content handler knows how do deal with a specific type of content on a wiki
  38  * page. Content is stored in the database in a serialized form (using a
  39  * serialization format a.k.a. MIME type) and is unserialized into its native
  40  * PHP representation (the content model), which is wrapped in an instance of
  41  * the appropriate subclass of Content.
  42  *
  43  * ContentHandler instances are stateless singletons that serve, among other
  44  * things, as a factory for Content objects. Generally, there is one subclass
  45  * of ContentHandler and one subclass of Content for every type of content model.
  46  *
  47  * Some content types have a flat model, that is, their native representation
  48  * is the same as their serialized form. Examples would be JavaScript and CSS
  49  * code. As of now, this also applies to wikitext (MediaWiki's default content
  50  * type), but wikitext content may be represented by a DOM or AST structure in
  51  * the future.
  52  *
  53  * @ingroup Content
  54  */
  55 abstract class ContentHandler {
  56         /**
  57          * Switch for enabling deprecation warnings. Used by ContentHandler::deprecated()
  58          * and ContentHandler::runLegacyHooks().
  59          *
  60          * Once the ContentHandler code has settled in a bit, this should be set to true to
  61          * make extensions etc. show warnings when using deprecated functions and hooks.
  62          */
  63         protected static $enableDeprecationWarnings = false;
  64
  65         /**
  66          * Convenience function for getting flat text from a Content object. This
  67          * should only be used in the context of backwards compatibility with code
  68          * that is not yet able to handle Content objects!
  69          *
  70          * If $content is null, this method returns the empty string.
  71          *
  72          * If $content is an instance of TextContent, this method returns the flat
  73          * text as returned by $content->getNativeData().
  74          *
  75          * If $content is not a TextContent object, the behavior of this method
  76          * depends on the global $wgContentHandlerTextFallback:
  77          * - If $wgContentHandlerTextFallback is 'fail' and $content is not a
  78          *   TextContent object, an MWException is thrown.
  79          * - If $wgContentHandlerTextFallback is 'serialize' and $content is not a
  80          *   TextContent object, $content->serialize() is called to get a string
  81          *   form of the content.
  82          * - If $wgContentHandlerTextFallback is 'ignore' and $content is not a
  83          *   TextContent object, this method returns null.
  84          * - otherwise, the behavior is undefined.
  85          *
  86          * @since 1.21
  87          *
  88          * @param Content $content
  89          *
  90          * @throws MWException If the content is not an instance of TextContent and
  91          * wgContentHandlerTextFallback was set to 'fail'.
  92          * @return string|null Textual form of the content, if available.
  93          */
  94         public static function getContentText( Content $content = null ) {
  95                 global $wgContentHandlerTextFallback;
  96
  97                 if ( is_null( $content ) ) {
  98                         return '';
  99                 }
 100
 101                 if ( $content instanceof TextContent ) {
 102                         return $content->getNativeData();
 103                 }
 104
 105                 wfDebugLog( 'ContentHandler', 'Accessing ' . $content->getModel() . ' content as text!' );
 106
 107                 if ( $wgContentHandlerTextFallback == 'fail' ) {
 108                         throw new MWException(
 109                                 "Attempt to get text from Content with model " .
 110                                 $content->getModel()
 111                         );
 112                 }
 113
 114                 if ( $wgContentHandlerTextFallback == 'serialize' ) {
 115                         return $content->serialize();
 116                 }
 117
 118                 return null;
 119         }
 120
 121         /**
 122          * Convenience function for creating a Content object from a given textual
 123          * representation.
 124          *
 125          * $text will be deserialized into a Content object of the model specified
 126          * by $modelId (or, if that is not given, $title->getContentModel()) using
 127          * the given format.
 128          *
 129          * @since 1.21
 130          *
 131          * @param string $text The textual representation, will be
 132          *    unserialized to create the Content object
 133          * @param Title $title The title of the page this text belongs to.
 134          *    Required if $modelId is not provided.
 135          * @param string $modelId The model to deserialize to. If not provided,
 136          *    $title->getContentModel() is used.
 137          * @param string $format The format to use for deserialization. If not
 138          *    given, the model's default format is used.
 139          *
 140          * @throws MWException If model ID or format is not supported or if the text can not be
 141          * unserialized using the format.
 142          * @return Content A Content object representing the text.
 143          */
 144         public static function makeContent( $text, Title $title = null,
 145                 $modelId = null, $format = null ) {
 146                 if ( is_null( $modelId ) ) {
 147                         if ( is_null( $title ) ) {
 148                                 throw new MWException( "Must provide a Title object or a content model ID." );
 149                         }
 150
 151                         $modelId = $title->getContentModel();
 152                 }
 153
 154                 $handler = ContentHandler::getForModelID( $modelId );
 155
 156                 return $handler->unserializeContent( $text, $format );
 157         }
 158
 159         /**
 160          * Returns the name of the default content model to be used for the page
 161          * with the given title.
 162          *
 163          * Note: There should rarely be need to call this method directly.
 164          * To determine the actual content model for a given page, use
 165          * Title::getContentModel().
 166          *
 167          * Which model is to be used by default for the page is determined based
 168          * on several factors:
 169          * - The global setting $wgNamespaceContentModels specifies a content model
 170          *   per namespace.
 171          * - The hook ContentHandlerDefaultModelFor may be used to override the page's default
 172          *   model.
 173          * - Pages in NS_MEDIAWIKI and NS_USER default to the CSS or JavaScript
 174          *   model if they end in .js or .css, respectively.
 175          * - Pages in NS_MEDIAWIKI default to the wikitext model otherwise.
 176          * - The hook TitleIsCssOrJsPage may be used to force a page to use the CSS
 177          *   or JavaScript model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 178          *   hook should be used instead if possible.
 179          * - The hook TitleIsWikitextPage may be used to force a page to use the
 180          *   wikitext model. This is a compatibility feature. The ContentHandlerDefaultModelFor
 181          *   hook should be used instead if possible.
 182          *
 183          * If none of the above applies, the wikitext model is used.
 184          *
 185          * Note: this is used by, and may thus not use, Title::getContentModel()
 186          *
 187          * @since 1.21
 188          *
 189          * @param Title $title
 190          *
 191          * @return string Default model name for the page given by $title
 192          */
 193         public static function getDefaultModelFor( Title $title ) {
 194                 // NOTE: this method must not rely on $title->getContentModel() directly or indirectly,
 195                 //       because it is used to initialize the mContentModel member.
 196
 197                 $ns = $title->getNamespace();
 198
 199                 $ext = false;
 200                 $m = null;
 201                 $model = MWNamespace::getNamespaceContentModel( $ns );
 202
 203                 // Hook can determine default model
 204                 if ( !wfRunHooks( 'ContentHandlerDefaultModelFor', array( $title, &$model ) ) ) {
 205                         if ( !is_null( $model ) ) {
 206                                 return $model;
 207                         }
 208                 }
 209
 210                 // Could this page contain custom CSS or JavaScript, based on the title?
 211                 $isCssOrJsPage = NS_MEDIAWIKI == $ns && preg_match( '!\.(css|js)$!u', $title->getText(), $m );
 212                 if ( $isCssOrJsPage ) {
 213                         $ext = $m[1];
 214                 }
 215
 216                 // Hook can force JS/CSS
 217                 wfRunHooks( 'TitleIsCssOrJsPage', array( $title, &$isCssOrJsPage ) );
 218
 219                 // Is this a .css subpage of a user page?
 220                 $isJsCssSubpage = NS_USER == $ns
 221                         && !$isCssOrJsPage
 222                         && preg_match( "/\\/.*\\.(js|css)$/", $title->getText(), $m );
 223                 if ( $isJsCssSubpage ) {
 224                         $ext = $m[1];
 225                 }
 226
 227                 // Is this wikitext, according to $wgNamespaceContentModels or the DefaultModelFor hook?
 228                 $isWikitext = is_null( $model ) || $model == CONTENT_MODEL_WIKITEXT;
 229                 $isWikitext = $isWikitext && !$isCssOrJsPage && !$isJsCssSubpage;
 230
 231                 // Hook can override $isWikitext
 232                 wfRunHooks( 'TitleIsWikitextPage', array( $title, &$isWikitext ) );
 233
 234                 if ( !$isWikitext ) {
 235                         switch ( $ext ) {
 236                                 case 'js':
 237                                         return CONTENT_MODEL_JAVASCRIPT;
 238                                 case 'css':
 239                                         return CONTENT_MODEL_CSS;
 240                                 default:
 241                                         return is_null( $model ) ? CONTENT_MODEL_TEXT : $model;
 242                         }
 243                 }
 244
 245                 // We established that it must be wikitext
 246
 247                 return CONTENT_MODEL_WIKITEXT;
 248         }
 249
 250         /**
 251          * Returns the appropriate ContentHandler singleton for the given title.
 252          *
 253          * @since 1.21
 254          *
 255          * @param Title $title
 256          *
 257          * @return ContentHandler
 258          */
 259         public static function getForTitle( Title $title ) {
 260                 $modelId = $title->getContentModel();
 261
 262                 return ContentHandler::getForModelID( $modelId );
 263         }
 264
 265         /**
 266          * Returns the appropriate ContentHandler singleton for the given Content
 267          * object.
 268          *
 269          * @since 1.21
 270          *
 271          * @param Content $content
 272          *
 273          * @return ContentHandler
 274          */
 275         public static function getForContent( Content $content ) {
 276                 $modelId = $content->getModel();
 277
 278                 return ContentHandler::getForModelID( $modelId );
 279         }
 280
 281         /**
 282          * @var array A Cache of ContentHandler instances by model id
 283          */
 284         protected static $handlers;
 285
 286         /**
 287          * Returns the ContentHandler singleton for the given model ID. Use the
 288          * CONTENT_MODEL_XXX constants to identify the desired content model.
 289          *
 290          * ContentHandler singletons are taken from the global $wgContentHandlers
 291          * array. Keys in that array are model names, the values are either
 292          * ContentHandler singleton objects, or strings specifying the appropriate
 293          * subclass of ContentHandler.
 294          *
 295          * If a class name is encountered when looking up the singleton for a given
 296          * model name, the class is instantiated and the class name is replaced by
 297          * the resulting singleton in $wgContentHandlers.
 298          *
 299          * If no ContentHandler is defined for the desired $modelId, the
 300          * ContentHandler may be provided by the ContentHandlerForModelID hook.
 301          * If no ContentHandler can be determined, an MWException is raised.
 302          *
 303          * @since 1.21
 304          *
 305          * @param string $modelId The ID of the content model for which to get a
 306          *    handler. Use CONTENT_MODEL_XXX constants.
 307          *
 308          * @throws MWException If no handler is known for the model ID.
 309          * @return ContentHandler The ContentHandler singleton for handling the model given by the ID.
 310          */
 311         public static function getForModelID( $modelId ) {
 312                 global $wgContentHandlers;
 313
 314                 if ( isset( ContentHandler::$handlers[$modelId] ) ) {
 315                         return ContentHandler::$handlers[$modelId];
 316                 }
 317
 318                 if ( empty( $wgContentHandlers[$modelId] ) ) {
 319                         $handler = null;
 320
 321                         wfRunHooks( 'ContentHandlerForModelID', array( $modelId, &$handler ) );
 322
 323                         if ( $handler === null ) {
 324                                 throw new MWException( "No handler for model '$modelId' registered in \$wgContentHandlers" );
 325                         }
 326
 327                         if ( !( $handler instanceof ContentHandler ) ) {
 328                                 throw new MWException( "ContentHandlerForModelID must supply a ContentHandler instance" );
 329                         }
 330                 } else {
 331                         $class = $wgContentHandlers[$modelId];
 332                         $handler = new $class( $modelId );
 333
 334                         if ( !( $handler instanceof ContentHandler ) ) {
 335                                 throw new MWException( "$class from \$wgContentHandlers is not " .
 336                                         "compatible with ContentHandler" );
 337                         }
 338                 }
 339
 340                 wfDebugLog( 'ContentHandler', 'Created handler for ' . $modelId
 341                         . ': ' . get_class( $handler ) );
 342
 343                 ContentHandler::$handlers[$modelId] = $handler;
 344
 345                 return ContentHandler::$handlers[$modelId];
 346         }
 347
 348         /**
 349          * Returns the localized name for a given content model.
 350          *
 351          * Model names are localized using system messages. Message keys
 352          * have the form content-model-$name, where $name is getContentModelName( $id ).
 353          *
 354          * @param string $name The content model ID, as given by a CONTENT_MODEL_XXX
 355          *    constant or returned by Revision::getContentModel().
 356          *
 357          * @throws MWException If the model ID isn't known.
 358          * @return string The content model's localized name.
 359          */
 360         public static function getLocalizedName( $name ) {
 361                 // Messages: content-model-wikitext, content-model-text,
 362                 // content-model-javascript, content-model-css
 363                 $key = "content-model-$name";
 364
 365                 $msg = wfMessage( $key );
 366
 367                 return $msg->exists() ? $msg->plain() : $name;
 368         }
 369
 370         public static function getContentModels() {
 371                 global $wgContentHandlers;
 372
 373                 return array_keys( $wgContentHandlers );
 374         }
 375
 376         public static function getAllContentFormats() {
 377                 global $wgContentHandlers;
 378
 379                 $formats = array();
 380
 381                 foreach ( $wgContentHandlers as $model => $class ) {
 382                         $handler = ContentHandler::getForModelID( $model );
 383                         $formats = array_merge( $formats, $handler->getSupportedFormats() );
 384                 }
 385
 386                 $formats = array_unique( $formats );
 387
 388                 return $formats;
 389         }
 390
 391         // ------------------------------------------------------------------------
 392
 393         /**
 394          * @var string
 395          */
 396         protected $mModelID;
 397
 398         /**
 399          * @var string[]
 400          */
 401         protected $mSupportedFormats;
 402
 403         /**
 404          * Constructor, initializing the ContentHandler instance with its model ID
 405          * and a list of supported formats. Values for the parameters are typically
 406          * provided as literals by subclass's constructors.
 407          *
 408          * @param string $modelId (use CONTENT_MODEL_XXX constants).
 409          * @param string[] $formats List for supported serialization formats
 410          *    (typically as MIME types)
 411          */
 412         public function __construct( $modelId, $formats ) {
 413                 $this->mModelID = $modelId;
 414                 $this->mSupportedFormats = $formats;
 415
 416                 $this->mModelName = preg_replace( '/(Content)?Handler$/', '', get_class( $this ) );
 417                 $this->mModelName = preg_replace( '/[_\\\\]/', '', $this->mModelName );
 418                 $this->mModelName = strtolower( $this->mModelName );
 419         }
 420
 421         /**
 422          * Serializes a Content object of the type supported by this ContentHandler.
 423          *
 424          * @since 1.21
 425          *
 426          * @param Content $content The Content object to serialize
 427          * @param string $format The desired serialization format
 428          *
 429          * @return string Serialized form of the content
 430          */
 431         abstract public function serializeContent( Content $content, $format = null );
 432
 433         /**
 434          * Applies transformations on export (returns the blob unchanged per default).
 435          * Subclasses may override this to perform transformations such as conversion
 436          * of legacy formats or filtering of internal meta-data.
 437          *
 438          * @param string $blob The blob to be exported
 439          * @param string|null $format The blob's serialization format
 440          *
 441          * @return string
 442          */
 443         public function exportTransform( $blob, $format = null ) {
 444                 return $blob;
 445         }
 446
 447         /**
 448          * Unserializes a Content object of the type supported by this ContentHandler.
 449          *
 450          * @since 1.21
 451          *
 452          * @param string $blob Serialized form of the content
 453          * @param string $format The format used for serialization
 454          *
 455          * @return Content The Content object created by deserializing $blob
 456          */
 457         abstract public function unserializeContent( $blob, $format = null );
 458
 459         /**
 460          * Apply import transformation (per default, returns $blob unchanged).
 461          * This gives subclasses an opportunity to transform data blobs on import.
 462          *
 463          * @singe 1.24
 464          *
 465          * @param string $blob
 466          * @param string|null $format
 467          *
 468          * @return string
 469          */
 470         public function importTransform( $blob, $format = null ) {
 471                 return $blob;
 472         }
 473
 474         /**
 475          * Creates an empty Content object of the type supported by this
 476          * ContentHandler.
 477          *
 478          * @since 1.21
 479          *
 480          * @return Content
 481          */
 482         abstract public function makeEmptyContent();
 483
 484         /**
 485          * Creates a new Content object that acts as a redirect to the given page,
 486          * or null if redirects are not supported by this content model.
 487          *
 488          * This default implementation always returns null. Subclasses supporting redirects
 489          * must override this method.
 490          *
 491          * Note that subclasses that override this method to return a Content object
 492          * should also override supportsRedirects() to return true.
 493          *
 494          * @since 1.21
 495          *
 496          * @param Title $destination The page to redirect to.
 497          * @param string $text Text to include in the redirect, if possible.
 498          *
 499          * @return Content Always null.
 500          */
 501         public function makeRedirectContent( Title $destination, $text = '' ) {
 502                 return null;
 503         }
 504
 505         /**
 506          * Returns the model id that identifies the content model this
 507          * ContentHandler can handle. Use with the CONTENT_MODEL_XXX constants.
 508          *
 509          * @since 1.21
 510          *
 511          * @return string The model ID
 512          */
 513         public function getModelID() {
 514                 return $this->mModelID;
 515         }
 516
 517         /**
 518          * @since 1.21
 519          *
 520          * @param string $model_id The model to check
 521          *
 522          * @throws MWException If the model ID is not the ID of the content model supported by this
 523          * ContentHandler.
 524          */
 525         protected function checkModelID( $model_id ) {
 526                 if ( $model_id !== $this->mModelID ) {
 527                         throw new MWException( "Bad content model: " .
 528                                 "expected {$this->mModelID} " .
 529                                 "but got $model_id." );
 530                 }
 531         }
 532
 533         /**
 534          * Returns a list of serialization formats supported by the
 535          * serializeContent() and unserializeContent() methods of this
 536          * ContentHandler.
 537          *
 538          * @since 1.21
 539          *
 540          * @return string[] List of serialization formats as MIME type like strings
 541          */
 542         public function getSupportedFormats() {
 543                 return $this->mSupportedFormats;
 544         }
 545
 546         /**
 547          * The format used for serialization/deserialization by default by this
 548          * ContentHandler.
 549          *
 550          * This default implementation will return the first element of the array
 551          * of formats that was passed to the constructor.
 552          *
 553          * @since 1.21
 554          *
 555          * @return string The name of the default serialization format as a MIME type
 556          */
 557         public function getDefaultFormat() {
 558                 return $this->mSupportedFormats[0];
 559         }
 560
 561         /**
 562          * Returns true if $format is a serialization format supported by this
 563          * ContentHandler, and false otherwise.
 564          *
 565          * Note that if $format is null, this method always returns true, because
 566          * null means "use the default format".
 567          *
 568          * @since 1.21
 569          *
 570          * @param string $format The serialization format to check
 571          *
 572          * @return bool
 573          */
 574         public function isSupportedFormat( $format ) {
 575                 if ( !$format ) {
 576                         return true; // this means "use the default"
 577                 }
 578
 579                 return in_array( $format, $this->mSupportedFormats );
 580         }
 581
 582         /**
 583          * Convenient for checking whether a format provided as a parameter is actually supported.
 584          *
 585          * @param string $format The serialization format to check
 586          *
 587          * @throws MWException If the format is not supported by this content handler.
 588          */
 589         protected function checkFormat( $format ) {
 590                 if ( !$this->isSupportedFormat( $format ) ) {
 591                         throw new MWException(
 592                                 "Format $format is not supported for content model "
 593                                 . $this->getModelID()
 594                         );
 595                 }
 596         }
 597
 598         /**
 599          * Returns overrides for action handlers.
 600          * Classes listed here will be used instead of the default one when
 601          * (and only when) $wgActions[$action] === true. This allows subclasses
 602          * to override the default action handlers.
 603          *
 604          * @since 1.21
 605          *
 606          * @return array Always an empty array.
 607          */
 608         public function getActionOverrides() {
 609                 return array();
 610         }
 611
 612         /**
 613          * Factory for creating an appropriate DifferenceEngine for this content model.
 614          *
 615          * @since 1.21
 616          *
 617          * @param IContextSource $context Context to use, anything else will be ignored.
 618          * @param int $old Revision ID we want to show and diff with.
 619          * @param int|string $new Either a revision ID or one of the strings 'cur', 'prev' or 'next'.
 620          * @param int $rcid FIXME: Deprecated, no longer used. Defaults to 0.
 621          * @param bool $refreshCache If set, refreshes the diff cache. Defaults to false.
 622          * @param bool $unhide If set, allow viewing deleted revs. Defaults to false.
 623          *
 624          * @return DifferenceEngine
 625          */
 626         public function createDifferenceEngine( IContextSource $context, $old = 0, $new = 0,
 627                 $rcid = 0, //FIXME: Deprecated, no longer used
 628                 $refreshCache = false, $unhide = false ) {
 629                 $diffEngineClass = $this->getDiffEngineClass();
 630
 631                 return new $diffEngineClass( $context, $old, $new, $rcid, $refreshCache, $unhide );
 632         }
 633
 634         /**
 635          * Get the language in which the content of the given page is written.
 636          *
 637          * This default implementation just returns $wgContLang (except for pages
 638          * in the MediaWiki namespace)
 639          *
 640          * Note that the pages language is not cacheable, since it may in some
 641          * cases depend on user settings.
 642          *
 643          * Also note that the page language may or may not depend on the actual content of the page,
 644          * that is, this method may load the content in order to determine the language.
 645          *
 646          * @since 1.21
 647          *
 648          * @param Title $title The page to determine the language for.
 649          * @param Content $content The page's content, if you have it handy, to avoid reloading it.
 650          *
 651          * @return Language The page's language
 652          */
 653         public function getPageLanguage( Title $title, Content $content = null ) {
 654                 global $wgContLang, $wgLang;
 655                 $pageLang = $wgContLang;
 656
 657                 if ( $title->getNamespace() == NS_MEDIAWIKI ) {
 658                         // Parse mediawiki messages with correct target language
 659                         list( /* $unused */, $lang ) = MessageCache::singleton()->figureMessage( $title->getText() );
 660                         $pageLang = wfGetLangObj( $lang );
 661                 }
 662
 663                 wfRunHooks( 'PageContentLanguage', array( $title, &$pageLang, $wgLang ) );
 664
 665                 return wfGetLangObj( $pageLang );
 666         }
 667
 668         /**
 669          * Get the language in which the content of this page is written when
 670          * viewed by user. Defaults to $this->getPageLanguage(), but if the user
 671          * specified a preferred variant, the variant will be used.
 672          *
 673          * This default implementation just returns $this->getPageLanguage( $title, $content ) unless
 674          * the user specified a preferred variant.
 675          *
 676          * Note that the pages view language is not cacheable, since it depends on user settings.
 677          *
 678          * Also note that the page language may or may not depend on the actual content of the page,
 679          * that is, this method may load the content in order to determine the language.
 680          *
 681          * @since 1.21
 682          *
 683          * @param Title $title The page to determine the language for.
 684          * @param Content $content The page's content, if you have it handy, to avoid reloading it.
 685          *
 686          * @return Language The page's language for viewing
 687          */
 688         public function getPageViewLanguage( Title $title, Content $content = null ) {
 689                 $pageLang = $this->getPageLanguage( $title, $content );
 690
 691                 if ( $title->getNamespace() !== NS_MEDIAWIKI ) {
 692                         // If the user chooses a variant, the content is actually
 693                         // in a language whose code is the variant code.
 694                         $variant = $pageLang->getPreferredVariant();
 695                         if ( $pageLang->getCode() !== $variant ) {
 696                                 $pageLang = Language::factory( $variant );
 697                         }
 698                 }
 699
 700                 return $pageLang;
 701         }
 702
 703         /**
 704          * Determines whether the content type handled by this ContentHandler
 705          * can be used on the given page.
 706          *
 707          * This default implementation always returns true.
 708          * Subclasses may override this to restrict the use of this content model to specific locations,
 709          * typically based on the namespace or some other aspect of the title, such as a special suffix
 710          * (e.g. ".svg" for SVG content).
 711          *
 712          * @note: this calls the ContentHandlerCanBeUsedOn hook which may be used to override which
 713          * content model can be used where.
 714          *
 715          * @param Title $title The page's title.
 716          *
 717          * @return bool True if content of this kind can be used on the given page, false otherwise.
 718          */
 719         public function canBeUsedOn( Title $title ) {
 720                 $ok = true;
 721
 722                 wfRunHooks( 'ContentModelCanBeUsedOn', array( $this->getModelID(), $title, &$ok ) );
 723
 724                 return $ok;
 725         }
 726
 727         /**
 728          * Returns the name of the diff engine to use.
 729          *
 730          * @since 1.21
 731          *
 732          * @return string
 733          */
 734         protected function getDiffEngineClass() {
 735                 return 'DifferenceEngine';
 736         }
 737
 738         /**
 739          * Attempts to merge differences between three versions. Returns a new
 740          * Content object for a clean merge and false for failure or a conflict.
 741          *
 742          * This default implementation always returns false.
 743          *
 744          * @since 1.21
 745          *
 746          * @param Content|string $oldContent The page's previous content.
 747          * @param Content|string $myContent One of the page's conflicting contents.
 748          * @param Content|string $yourContent One of the page's conflicting contents.
 749          *
 750          * @return Content|bool Always false.
 751          */
 752         public function merge3( Content $oldContent, Content $myContent, Content $yourContent ) {
 753                 return false;
 754         }
 755
 756         /**
 757          * Return an applicable auto-summary if one exists for the given edit.
 758          *
 759          * @since 1.21
 760          *
 761          * @param Content $oldContent The previous text of the page.
 762          * @param Content $newContent The submitted text of the page.
 763          * @param int $flags Bit mask: a bit mask of flags submitted for the edit.
 764          *
 765          * @return string An appropriate auto-summary, or an empty string.
 766          */
 767         public function getAutosummary( Content $oldContent = null, Content $newContent = null,
 768                 $flags ) {
 769                 // Decide what kind of auto-summary is needed.
 770
 771                 // Redirect auto-summaries
 772
 773                 /**
 774                  * @var $ot Title
 775                  * @var $rt Title
 776                  */
 777
 778                 $ot = !is_null( $oldContent ) ? $oldContent->getRedirectTarget() : null;
 779                 $rt = !is_null( $newContent ) ? $newContent->getRedirectTarget() : null;
 780
 781                 if ( is_object( $rt ) ) {
 782                         if ( !is_object( $ot )
 783                                 || !$rt->equals( $ot )
 784                                 || $ot->getFragment() != $rt->getFragment()
 785                         ) {
 786                                 $truncatedtext = $newContent->getTextForSummary(
 787                                         250
 788                                         - strlen( wfMessage( 'autoredircomment' )->inContentLanguage()->text() )
 789                                         - strlen( $rt->getFullText() ) );
 790
 791                                 return wfMessage( 'autoredircomment', $rt->getFullText() )
 792                                         ->rawParams( $truncatedtext )->inContentLanguage()->text();
 793                         }
 794                 }
 795
 796                 // New page auto-summaries
 797                 if ( $flags & EDIT_NEW && $newContent->getSize() > 0 ) {
 798                         // If they're making a new article, give its text, truncated, in
 799                         // the summary.
 800
 801                         $truncatedtext = $newContent->getTextForSummary(
 802                                 200 - strlen( wfMessage( 'autosumm-new' )->inContentLanguage()->text() ) );
 803
 804                         return wfMessage( 'autosumm-new' )->rawParams( $truncatedtext )
 805                                 ->inContentLanguage()->text();
 806                 }
 807
 808                 // Blanking auto-summaries
 809                 if ( !empty( $oldContent ) && $oldContent->getSize() > 0 && $newContent->getSize() == 0 ) {
 810                         return wfMessage( 'autosumm-blank' )->inContentLanguage()->text();
 811                 } elseif ( !empty( $oldContent )
 812                         && $oldContent->getSize() > 10 * $newContent->getSize()
 813                         && $newContent->getSize() < 500
 814                 ) {
 815                         // Removing more than 90% of the article
 816
 817                         $truncatedtext = $newContent->getTextForSummary(
 818                                 200 - strlen( wfMessage( 'autosumm-replace' )->inContentLanguage()->text() ) );
 819
 820                         return wfMessage( 'autosumm-replace' )->rawParams( $truncatedtext )
 821                                 ->inContentLanguage()->text();
 822                 }
 823
 824                 // If we reach this point, there's no applicable auto-summary for our
 825                 // case, so our auto-summary is empty.
 826                 return '';
 827         }
 828
 829         /**
 830          * Auto-generates a deletion reason
 831          *
 832          * @since 1.21
 833          *
 834          * @param Title $title The page's title
 835          * @param bool &$hasHistory Whether the page has a history
 836          *
 837          * @return mixed String containing deletion reason or empty string, or
 838          *    boolean false if no revision occurred
 839          *
 840          * @XXX &$hasHistory is extremely ugly, it's here because
 841          * WikiPage::getAutoDeleteReason() and Article::generateReason()
 842          * have it / want it.
 843          */
 844         public function getAutoDeleteReason( Title $title, &$hasHistory ) {
 845                 $dbw = wfGetDB( DB_MASTER );
 846
 847                 // Get the last revision
 848                 $rev = Revision::newFromTitle( $title );
 849
 850                 if ( is_null( $rev ) ) {
 851                         return false;
 852                 }
 853
 854                 // Get the article's contents
 855                 $content = $rev->getContent();
 856                 $blank = false;
 857
 858                 // If the page is blank, use the text from the previous revision,
 859                 // which can only be blank if there's a move/import/protect dummy
 860                 // revision involved
 861                 if ( !$content || $content->isEmpty() ) {
 862                         $prev = $rev->getPrevious();
 863
 864                         if ( $prev ) {
 865                                 $rev = $prev;
 866                                 $content = $rev->getContent();
 867                                 $blank = true;
 868                         }
 869                 }
 870
 871                 $this->checkModelID( $rev->getContentModel() );
 872
 873                 // Find out if there was only one contributor
 874                 // Only scan the last 20 revisions
 875                 $res = $dbw->select( 'revision', 'rev_user_text',
 876                         array(
 877                                 'rev_page' => $title->getArticleID(),
 878                                 $dbw->bitAnd( 'rev_deleted', Revision::DELETED_USER ) . ' = 0'
 879                         ),
 880                         __METHOD__,
 881                         array( 'LIMIT' => 20 )
 882                 );
 883
 884                 if ( $res === false ) {
 885                         // This page has no revisions, which is very weird
 886                         return false;
 887                 }
 888
 889                 $hasHistory = ( $res->numRows() > 1 );
 890                 $row = $dbw->fetchObject( $res );
 891
 892                 if ( $row ) { // $row is false if the only contributor is hidden
 893                         $onlyAuthor = $row->rev_user_text;
 894                         // Try to find a second contributor
 895                         foreach ( $res as $row ) {
 896                                 if ( $row->rev_user_text != $onlyAuthor ) { // Bug 22999
 897                                         $onlyAuthor = false;
 898                                         break;
 899                                 }
 900                         }
 901                 } else {
 902                         $onlyAuthor = false;
 903                 }
 904
 905                 // Generate the summary with a '$1' placeholder
 906                 if ( $blank ) {
 907                         // The current revision is blank and the one before is also
 908                         // blank. It's just not our lucky day
 909                         $reason = wfMessage( 'exbeforeblank', '$1' )->inContentLanguage()->text();
 910                 } else {
 911                         if ( $onlyAuthor ) {
 912                                 $reason = wfMessage(
 913                                         'excontentauthor',
 914                                         '$1',
 915                                         $onlyAuthor
 916                                 )->inContentLanguage()->text();
 917                         } else {
 918                                 $reason = wfMessage( 'excontent', '$1' )->inContentLanguage()->text();
 919                         }
 920                 }
 921
 922                 if ( $reason == '-' ) {
 923                         // Allow these UI messages to be blanked out cleanly
 924                         return '';
 925                 }
 926
 927                 // Max content length = max comment length - length of the comment (excl. $1)
 928                 $text = $content ? $content->getTextForSummary( 255 - ( strlen( $reason ) - 2 ) ) : '';
 929
 930                 // Now replace the '$1' placeholder
 931                 $reason = str_replace( '$1', $text, $reason );
 932
 933                 return $reason;
 934         }
 935
 936         /**
 937          * Get the Content object that needs to be saved in order to undo all revisions
 938          * between $undo and $undoafter. Revisions must belong to the same page,
 939          * must exist and must not be deleted.
 940          *
 941          * @since 1.21
 942          *
 943          * @param Revision $current The current text
 944          * @param Revision $undo The revision to undo
 945          * @param Revision $undoafter Must be an earlier revision than $undo
 946          *
 947          * @return mixed String on success, false on failure
 948          */
 949         public function getUndoContent( Revision $current, Revision $undo, Revision $undoafter ) {
 950                 $cur_content = $current->getContent();
 951
 952                 if ( empty( $cur_content ) ) {
 953                         return false; // no page
 954                 }
 955
 956                 $undo_content = $undo->getContent();
 957                 $undoafter_content = $undoafter->getContent();
 958
 959                 if ( !$undo_content || !$undoafter_content ) {
 960                         return false; // no content to undo
 961                 }
 962
 963                 $this->checkModelID( $cur_content->getModel() );
 964                 $this->checkModelID( $undo_content->getModel() );
 965                 $this->checkModelID( $undoafter_content->getModel() );
 966
 967                 if ( $cur_content->equals( $undo_content ) ) {
 968                         // No use doing a merge if it's just a straight revert.
 969                         return $undoafter_content;
 970                 }
 971
 972                 $undone_content = $this->merge3( $undo_content, $undoafter_content, $cur_content );
 973
 974                 return $undone_content;
 975         }
 976
 977         /**
 978          * Get parser options suitable for rendering and caching the article
 979          *
 980          * @param IContextSource|User|string $context One of the following:
 981          *        - IContextSource: Use the User and the Language of the provided
 982          *                                            context
 983          *        - User: Use the provided User object and $wgLang for the language,
 984          *                                            so use an IContextSource object if possible.
 985          *        - 'canonical': Canonical options (anonymous user with default
 986          *                                            preferences and content language).
 987          *
 988          * @throws MWException
 989          * @return ParserOptions
 990          */
 991         public function makeParserOptions( $context ) {
 992                 global $wgContLang, $wgEnableParserLimitReporting;
 993
 994                 if ( $context instanceof IContextSource ) {
 995                         $options = ParserOptions::newFromContext( $context );
 996                 } elseif ( $context instanceof User ) { // settings per user (even anons)
 997                         $options = ParserOptions::newFromUser( $context );
 998                 } elseif ( $context === 'canonical' ) { // canonical settings
 999                         $options = ParserOptions::newFromUserAndLang( new User, $wgContLang );
1000                 } else {
1001                         throw new MWException( "Bad context for parser options: $context" );
1002                 }
1003
1004                 $options->enableLimitReport( $wgEnableParserLimitReporting ); // show inclusion/loop reports
1005                 $options->setTidy( true ); // fix bad HTML
1006
1007                 return $options;
1008         }
1009
1010         /**
1011          * Returns true for content models that support caching using the
1012          * ParserCache mechanism. See WikiPage::isParserCacheUsed().
1013          *
1014          * @since 1.21
1015          *
1016          * @return bool Always false.
1017          */
1018         public function isParserCacheSupported() {
1019                 return false;
1020         }
1021
1022         /**
1023          * Returns true if this content model supports sections.
1024          * This default implementation returns false.
1025          *
1026          * Content models that return true here should also implement
1027          * Content::getSection, Content::replaceSection, etc. to handle sections..
1028          *
1029          * @return bool Always false.
1030          */
1031         public function supportsSections() {
1032                 return false;
1033         }
1034
1035         /**
1036          * Returns true if this content model supports redirects.
1037          * This default implementation returns false.
1038          *
1039          * Content models that return true here should also implement
1040          * ContentHandler::makeRedirectContent to return a Content object.
1041          *
1042          * @return bool Always false.
1043          */
1044         public function supportsRedirects() {
1045                 return false;
1046         }
1047
1048         /**
1049          * Logs a deprecation warning, visible if $wgDevelopmentWarnings, but only if
1050          * self::$enableDeprecationWarnings is set to true.
1051          *
1052          * @param string $func The name of the deprecated function
1053          * @param string $version The version since the method is deprecated. Usually 1.21
1054          *   for ContentHandler related stuff.
1055          * @param string|bool $component : Component to which the function belongs.
1056          *   If false, it is assumed the function is in MediaWiki core.
1057          *
1058          * @see ContentHandler::$enableDeprecationWarnings
1059          * @see wfDeprecated
1060          */
1061         public static function deprecated( $func, $version, $component = false ) {
1062                 if ( self::$enableDeprecationWarnings ) {
1063                         wfDeprecated( $func, $version, $component, 3 );
1064                 }
1065         }
1066
1067         /**
1068          * Call a legacy hook that uses text instead of Content objects.
1069          * Will log a warning when a matching hook function is registered.
1070          * If the textual representation of the content is changed by the
1071          * hook function, a new Content object is constructed from the new
1072          * text.
1073          *
1074          * @param string $event Event name
1075          * @param array $args Parameters passed to hook functions
1076          * @param bool $warn Whether to log a warning.
1077          *                    Default to self::$enableDeprecationWarnings.
1078          *                    May be set to false for testing.
1079          *
1080          * @return bool True if no handler aborted the hook
1081          *
1082          * @see ContentHandler::$enableDeprecationWarnings
1083          */
1084         public static function runLegacyHooks( $event, $args = array(),
1085                 $warn = null
1086         ) {
1087
1088                 if ( $warn === null ) {
1089                         $warn = self::$enableDeprecationWarnings;
1090                 }
1091
1092                 if ( !Hooks::isRegistered( $event ) ) {
1093                         return true; // nothing to do here
1094                 }
1095
1096                 if ( $warn ) {
1097                         // Log information about which handlers are registered for the legacy hook,
1098                         // so we can find and fix them.
1099
1100                         $handlers = Hooks::getHandlers( $event );
1101                         $handlerInfo = array();
1102
1103                         wfSuppressWarnings();
1104
1105                         foreach ( $handlers as $handler ) {
1106                                 if ( is_array( $handler ) ) {
1107                                         if ( is_object( $handler[0] ) ) {
1108                                                 $info = get_class( $handler[0] );
1109                                         } else {
1110                                                 $info = $handler[0];
1111                                         }
1112
1113                                         if ( isset( $handler[1] ) ) {
1114                                                 $info .= '::' . $handler[1];
1115                                         }
1116                                 } elseif ( is_object( $handler ) ) {
1117                                         $info = get_class( $handler[0] );
1118                                         $info .= '::on' . $event;
1119                                 } else {
1120                                         $info = $handler;
1121                                 }
1122
1123                                 $handlerInfo[] = $info;
1124                         }
1125
1126                         wfRestoreWarnings();
1127
1128                         wfWarn( "Using obsolete hook $event via ContentHandler::runLegacyHooks()! Handlers: " .
1129                                 implode( ', ', $handlerInfo ), 2 );
1130                 }
1131
1132                 // convert Content objects to text
1133                 $contentObjects = array();
1134                 $contentTexts = array();
1135
1136                 foreach ( $args as $k => $v ) {
1137                         if ( $v instanceof Content ) {
1138                                 /* @var Content $v */
1139
1140                                 $contentObjects[$k] = $v;
1141
1142                                 $v = $v->serialize();
1143                                 $contentTexts[$k] = $v;
1144                                 $args[$k] = $v;
1145                         }
1146                 }
1147
1148                 // call the hook functions
1149                 $ok = wfRunHooks( $event, $args );
1150
1151                 // see if the hook changed the text
1152                 foreach ( $contentTexts as $k => $orig ) {
1153                         /* @var Content $content */
1154
1155                         $modified = $args[$k];
1156                         $content = $contentObjects[$k];
1157
1158                         if ( $modified !== $orig ) {
1159                                 // text was changed, create updated Content object
1160                                 $content = $content->getContentHandler()->unserializeContent( $modified );
1161                         }
1162
1163                         $args[$k] = $content;
1164                 }
1165
1166                 return $ok;
1167         }
1168 }